Introduction

The Mapping Police Violence database contains information on civilians killed in incidents where law enforcement officers applied lethal force. The dataset includes variables such as personal information (name, age, available image), socio-demographic details (gender, race), and incident outcomes (location of the incident, responsible agency (police department, sheriff’s office, etc.), cause of death, and officer charges).

In this report, I will conduct a crude rate analysis of civilians exhibiting signs of mental illness, stratified by state and gender. The results will be presented through a map visualizing these crude rates. Additionally, Fisher’s exact tests were performed across states to compare the prevalence of reported signs of mental illness between genders.

(https://airtable.com/appzVzSeINK1S3EVR/shroOenW19l1m3w0H/tblxearKzw8W7ViN8)

Data Load & Processing

Mapping Police Violence

here::i_am("PeterkinC_FinalProject_Report.Rmd")
## here() starts at C:/Users/cpeterk/OneDrive - Emory/Fall 2025/DATA 550/Final Project/Milestone 3 - Organization and GitHub repository
mpv_data <- readRDS(
  file = here::here("output/mpv_data.rds")
)

The Mapping Police Violence database has 14980 rows and 61 columns.

American Community Survey (ACS)

state_pop <- readRDS(
  file = here::here("output/state_pop.rds")
)

State population estimates stratified by gender were derived from the American Community Survey 1-year estimates. My created state_pop database has 104 rows and 5 columns.

Data Processing

Here in my data processing step, I combined the police violence data mpv_data with state population estimates from the ACS to calculate crude rates of civilians killed who exhibited signs of mental illness. Only responses of “Yes” were considered in the exhibited signs of mental illness instance count. Counts of victims were aggregated by state and gender, then divided by the respective population to create standardized rates per 100,000 residents.

mental_crude_state <- readRDS(
  file = here::here("output/mental_crude_state.rds")
)

Table

The stratified tables below show how the presence of reported signs of mental illness among civilians killed varies by gender across U.S. states. Each table displays counts and percentages for Male and Female victims separately, as well as a statistical test comparing the two groups.

It is worth noting, that in many states, the proportion of male victims exhibiting signs of mental illness is higher than that of females, although the strength of this difference isn’t well proven. The only state that observed a statistically significant gender difference and reinforced the idea that male victims disproportionately exhibit signs of mental illness is New Mexico.

Characteristic
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
District of Columbia
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
Female
N = 18
1
Male
N = 265
1
p-value2 Female
N = 4
1
Male
N = 79
1
p-value2 Female
N = 37
1
Male
N = 602
1
p-value2 Female
N = 8
1
Male
N = 199
1
p-value2 Female
N = 110
1
Male
N = 1,953
1
p-value2 Female
N = 20
1
Male
N = 453
1
p-value2 Female
N = 3
1
Male
N = 61
1
p-value2 Female
N = 3
1
Male
N = 32
1
p-value2 Female
N = 5
1
Male
N = 44
1
p-value2 Female
N = 75
1
Male
N = 976
1
p-value2 Female
N = 40
1
Male
N = 541
1
p-value2 Female
N = 3
1
Male
N = 61
1
p-value2 Female
N = 7
1
Male
N = 103
1
p-value2 Female
N = 24
1
Male
N = 322
1
p-value2 Female
N = 17
1
Male
N = 280
1
p-value2 Female
N = 5
1
Male
N = 93
1
p-value2 Female
N = 10
1
Male
N = 131
1
p-value2 Female
N = 11
1
Male
N = 231
1
p-value2 Female
N = 10
1
Male
N = 278
1
p-value2 Female
N = 3
1
Male
N = 66
1
p-value2 Female
N = 13
1
Male
N = 200
1
p-value2 Female
N = 5
1
Male
N = 89
1
p-value2 Female
N = 12
1
Male
N = 252
1
p-value2 Female
N = 6
1
Male
N = 139
1
p-value2 Female
N = 11
1
Male
N = 199
1
p-value2 Female
N = 22
1
Male
N = 365
1
p-value2 Female
N = 3
1
Male
N = 84
1
p-value2 Female
N = 4
1
Male
N = 67
1
p-value2 Female
N = 9
1
Male
N = 228
1
p-value2 Female
N = 1
1
Male
N = 39
1
p-value2 Female
N = 14
1
Male
N = 166
1
p-value2 Female
N = 21
1
Male
N = 277
1
p-value2 Female
N = 25
1
Male
N = 310
1
p-value2 Female
N = 27
1
Male
N = 408
1
p-value2 Female
N = 1
1
Male
N = 31
1
p-value2 Female
N = 26
1
Male
N = 385
1
p-value2 Female
N = 21
1
Male
N = 347
1
p-value2 Female
N = 7
1
Male
N = 208
1
p-value2 Female
N = 16
1
Male
N = 319
1
p-value2 Male
N = 10
1
p-value3 Female
N = 24
1
Male
N = 225
1
p-value2 Female
N = 1
1
Male
N = 46
1
p-value2 Female
N = 29
1
Male
N = 364
1
p-value2 Female
N = 94
1
Male
N = 1,348
1
p-value2 Female
N = 9
1
Male
N = 177
1
p-value2 Female
N = 1
1
Male
N = 21
1
p-value2 Female
N = 18
1
Male
N = 252
1
p-value2 Female
N = 15
1
Male
N = 370
1
p-value2 Female
N = 4
1
Male
N = 127
1
p-value2 Female
N = 9
1
Male
N = 204
1
p-value2 Female
N = 2
1
Male
N = 47
1
p-value2
signs_of_mental_illness

0.13

0.6

0.13

0.8

0.2

0.6

0.8

0.7

0.3

0.2

0.9

0.7

0.5

0.5

0.7

0.5

>0.9

0.9

0.8

>0.9

>0.9

>0.9

0.5

0.6

0.6

>0.9

0.8

>0.9

0.2

0.2

0.3

0.045

0.5

0.5

0.3

0.4

0.4

0.2

0.8



0.5

0.15

>0.9

0.4

0.2

0.091

0.15

0.4

0.9

0.15

>0.9
     3 (17%) 8 (3.0%)
0 (0%) 2 (2.5%)
2 (5.4%) 17 (2.8%)
0 (0%) 5 (2.5%)
2 (1.8%) 49 (2.5%)
0 (0%) 14 (3.1%)



0 (0%) 2 (6.3%)



2 (2.7%) 25 (2.6%)
1 (2.5%) 23 (4.3%)
0 (0%) 1 (1.6%)
0 (0%) 3 (2.9%)
1 (4.2%) 5 (1.6%)
0 (0%) 7 (2.5%)



0 (0%) 5 (3.8%)
0 (0%) 6 (2.6%)
0 (0%) 5 (1.8%)



0 (0%) 7 (3.5%)



1 (8.3%) 10 (4.0%)
0 (0%) 1 (0.7%)
0 (0%) 6 (3.0%)
0 (0%) 8 (2.2%)
0 (0%) 3 (3.6%)
0 (0%) 1 (1.5%)
0 (0%) 8 (3.5%)
0 (0%) 2 (5.1%)
1 (7.1%) 3 (1.8%)
1 (4.8%) 6 (2.2%)
1 (4.0%) 8 (2.6%)
1 (3.7%) 6 (1.5%)



1 (3.8%) 8 (2.1%)
0 (0%) 11 (3.2%)
0 (0%) 11 (5.3%)
0 (0%) 7 (2.2%)
1 (10%)
1 (4.2%) 6 (2.7%)
0 (0%) 1 (2.2%)
0 (0%) 10 (2.7%)
2 (2.1%) 49 (3.6%)
1 (11%) 2 (1.1%)
1 (100%) 0 (0%)
1 (5.6%) 3 (1.2%)
1 (6.7%) 11 (3.0%)
0 (0%) 4 (3.1%)
2 (22%) 6 (2.9%)
0 (0%) 1 (2.1%)
    Drug or Alcohol Use 0 (0%) 8 (3.0%)
0 (0%) 4 (5.1%)
1 (2.7%) 29 (4.8%)
0 (0%) 8 (4.0%)
3 (2.7%) 140 (7.2%)
0 (0%) 26 (5.7%)
0 (0%) 3 (4.9%)
0 (0%) 1 (3.1%)
0 (0%) 1 (2.3%)
1 (1.3%) 56 (5.7%)
1 (2.5%) 25 (4.6%)
0 (0%) 4 (6.6%)
0 (0%) 8 (7.8%)
0 (0%) 10 (3.1%)
1 (5.9%) 12 (4.3%)
0 (0%) 7 (7.5%)
0 (0%) 10 (7.6%)
1 (9.1%) 11 (4.8%)
0 (0%) 11 (4.0%)
0 (0%) 6 (9.1%)
0 (0%) 8 (4.0%)
0 (0%) 5 (5.6%)
0 (0%) 12 (4.8%)
0 (0%) 7 (5.0%)
0 (0%) 7 (3.5%)
0 (0%) 19 (5.2%)
0 (0%) 8 (9.5%)
0 (0%) 2 (3.0%)
0 (0%) 14 (6.1%)
0 (0%) 2 (5.1%)
0 (0%) 5 (3.0%)
2 (9.5%) 11 (4.0%)
0 (0%) 23 (7.4%)
2 (7.4%) 26 (6.4%)
0 (0%) 2 (6.5%)
0 (0%) 26 (6.8%)
0 (0%) 19 (5.5%)
1 (14%) 13 (6.3%)
1 (6.3%) 11 (3.4%)
1 (10%)
0 (0%) 9 (4.0%)
1 (100%) 5 (11%)
1 (3.4%) 15 (4.1%)
3 (3.2%) 63 (4.7%)
0 (0%) 9 (5.1%)
0 (0%) 1 (4.8%)
0 (0%) 7 (2.8%)
1 (6.7%) 29 (7.8%)
0 (0%) 7 (5.5%)
0 (0%) 7 (3.4%)
0 (0%) 6 (13%)
    No 10 (56%) 162 (61%)
2 (50%) 48 (61%)
23 (62%) 380 (63%)
5 (63%) 135 (68%)
59 (54%) 1,122 (57%)
14 (70%) 313 (69%)
3 (100%) 34 (56%)
1 (33%) 17 (53%)
2 (40%) 32 (73%)
41 (55%) 579 (59%)
27 (68%) 365 (67%)
3 (100%) 41 (67%)
3 (43%) 62 (60%)
14 (58%) 206 (64%)
9 (53%) 175 (63%)
2 (40%) 56 (60%)
7 (70%) 71 (54%)
8 (73%) 156 (68%)
8 (80%) 170 (61%)
2 (67%) 39 (59%)
10 (77%) 119 (60%)
3 (60%) 46 (52%)
5 (42%) 143 (57%)
3 (50%) 78 (56%)
6 (55%) 134 (67%)
15 (68%) 237 (65%)
2 (67%) 49 (58%)
3 (75%) 35 (52%)
3 (33%) 131 (57%)
0 (0%) 21 (54%)
11 (79%) 99 (60%)
10 (48%) 179 (65%)
14 (56%) 161 (52%)
13 (48%) 244 (60%)
0 (0%) 22 (71%)
14 (54%) 224 (58%)
17 (81%) 200 (58%)
1 (14%) 105 (50%)
11 (69%) 189 (59%)
7 (70%)
16 (67%) 142 (63%)
0 (0%) 24 (52%)
18 (62%) 231 (63%)
54 (57%) 843 (63%)
7 (78%) 100 (56%)
0 (0%) 11 (52%)
10 (56%) 154 (61%)
7 (47%) 197 (53%)
2 (50%) 72 (57%)
4 (44%) 119 (58%)
2 (100%) 25 (53%)
    Unknown 1 (5.6%) 32 (12%)
1 (25%) 12 (15%)
3 (8.1%) 76 (13%)
2 (25%) 25 (13%)
16 (15%) 258 (13%)
1 (5.0%) 37 (8.2%)
0 (0%) 10 (16%)
1 (33%) 5 (16%)
1 (20%) 5 (11%)
7 (9.3%) 103 (11%)
3 (7.5%) 52 (9.6%)



1 (14%) 12 (12%)
3 (13%) 46 (14%)
2 (12%) 23 (8.2%)
1 (20%) 7 (7.5%)
1 (10%) 14 (11%)
1 (9.1%) 31 (13%)
1 (10%) 43 (15%)
0 (0%) 5 (7.6%)
1 (7.7%) 19 (9.5%)
0 (0%) 12 (13%)
2 (17%) 27 (11%)
0 (0%) 15 (11%)
3 (27%) 32 (16%)
3 (14%) 41 (11%)
0 (0%) 12 (14%)
0 (0%) 11 (16%)
3 (33%) 24 (11%)
1 (100%) 3 (7.7%)
0 (0%) 20 (12%)
0 (0%) 28 (10%)
1 (4.0%) 28 (9.0%)
3 (11%) 40 (9.8%)
0 (0%) 2 (6.5%)
4 (15%) 40 (10%)
2 (9.5%) 42 (12%)
1 (14%) 24 (12%)
1 (6.3%) 38 (12%)


1 (4.2%) 29 (13%)
0 (0%) 7 (15%)
4 (14%) 37 (10%)
10 (11%) 148 (11%)
0 (0%) 16 (9.0%)
0 (0%) 3 (14%)
0 (0%) 29 (12%)
0 (0%) 37 (10%)
1 (25%) 23 (18%)
0 (0%) 15 (7.4%)
0 (0%) 10 (21%)
    Yes 4 (22%) 55 (21%)
1 (25%) 13 (16%)
7 (19%) 100 (17%)
1 (13%) 26 (13%)
30 (27%) 384 (20%)
5 (25%) 63 (14%)
0 (0%) 14 (23%)
1 (33%) 7 (22%)
2 (40%) 6 (14%)
24 (32%) 213 (22%)
8 (20%) 76 (14%)
0 (0%) 15 (25%)
3 (43%) 18 (17%)
6 (25%) 55 (17%)
5 (29%) 63 (23%)
2 (40%) 23 (25%)
2 (20%) 31 (24%)
1 (9.1%) 27 (12%)
1 (10%) 49 (18%)
1 (33%) 16 (24%)
2 (15%) 47 (24%)
2 (40%) 26 (29%)
4 (33%) 60 (24%)
3 (50%) 38 (27%)
2 (18%) 20 (10%)
4 (18%) 60 (16%)
1 (33%) 12 (14%)
1 (25%) 18 (27%)
3 (33%) 51 (22%)
0 (0%) 11 (28%)
2 (14%) 39 (23%)
8 (38%) 53 (19%)
9 (36%) 90 (29%)
8 (30%) 92 (23%)
1 (100%) 5 (16%)
7 (27%) 87 (23%)
2 (9.5%) 75 (22%)
4 (57%) 55 (26%)
3 (19%) 74 (23%)
1 (10%)
6 (25%) 39 (17%)
0 (0%) 9 (20%)
6 (21%) 71 (20%)
25 (27%) 245 (18%)
1 (11%) 50 (28%)
0 (0%) 6 (29%)
7 (39%) 59 (23%)
6 (40%) 96 (26%)
1 (25%) 21 (17%)
3 (33%) 57 (28%)
0 (0%) 5 (11%)
    History of Drug or Alcohol Use





1 (2.7%) 0 (0%)















































































































































1 n (%)
2 Fisher’s exact test
3 NA

Map

Below is an interactive choropleth map displaying the crude rates of mental illness signs by state, faceted by gender. Again, only responses of “Yes” were considered in the exhibited signs of mental illness instance count. By looking at the map, you can see males experience higher crude rates of exhibiting signs of mental illness in comparison to females.

map_one <- readRDS(
  file = here::here("output/map_one.rds")
)

girafe(ggobj = map_one)